Deformation behaviour and microstructures of semi-solid A356.2 alloy prepared by radial forging process during high solid fraction compression

Abstract

Radial forging was introduced to the strain-induced step in the strain-induced melt activation process to prepare high-quality semi-solid A356.2 billet for the high solid fraction compression. Then, the deformation behaviour and microstructures at different compression velocities, temperatures and deformation zones were investigated. The results showed that radial forging can induce enough strain at 60% reduction of area to prepare ideal semi-solid microstructure. The microstructure had no obvious improvement at 75% reduction of area because the distortion energy may be saturated at 60%. During compression tests, the flow stress was sensitive to compression velocity (V_c) but was insensitive to holding temperature (T_h), and it obeyed the power law $σ_{p} = K_{c} {\bar{\overset{\cdot}{ε}}}^{m}$ . The average strain rate sensitivity was 0.2757, and the average apparent activation energy was 254.5 kJ/mol. The compression sample can be divided into hard deformation zone, transition deformation zone, severe deformation zone and free deformation zone. Different zones had different predominant deformation mechanisms which not only affected the morphology of the grains but also codetermined the flow stress. With increasing V_c and T_h, the average particle size was increased from hard deformation zone, reached the maximum at severe deformation zone and then decreased at free deformation zone, while the opposite was the case for shape factor. At each deformation zone, the average particle size was increased while the shape factor was decreased with the increasing V_c because higher V_c caused more grains to merge into larger and more irregular grains, while the opposite was the case for the increasing T_h because higher T_h generated more liquid which helped to isolate the contacted particles and improve their spheroidization degree.

Keywords

Semi-solid A356.2 alloy strain-induced melt activation thixotropic deformation microstructure

Introduction

Semi-solid metal processing (SSMP) has attained considerable interest owing to its distinctive rheological characteristics that originated from spherical-like microstructures.^1,2 Compared with conventional casting and forging, SSMP exhibits lower forming temperature, less deformation resistance, better formability and higher mechanical properties.^3,4 Thus, investigations on the preparation technology, microstructural evolution and deformation behaviour of the semi-solid alloys have caused lots of attention.

Among different kinds of production technologies, strain-induced melt activation (SIMA) is one of the most widely investigated technologies to obtain fine and spheroidal solid particles,^5,6 which include deformation step and partial re-melting step. The former is obviously more crucial and caused more researches. Here, we summarize the common strain-induced methods used in the SIMA process as follows: cold compression and warm compression,^7,8 equal channel angular extrusion (ECAE),⁹ cyclic upsetting-extrusion process (CUEP),¹⁰ rolling,¹¹ cross wedge rolling (CWR),¹² multi-axial forging (MAF),¹³ accumulative back extrusion (ABE),¹⁴ cyclic closed-die forging (CCDF)¹⁵ and cyclic extrusion compression (CEC).¹⁶ Although high-quality semi-solid billet can also be prepared by the above-mentioned method, the billet dimensions are severely limited by the capacity of the equipment. However, the investigation on the large semi-solid billet preparation in SIMA technologies for industrial application is less presented.

Furthermore, many investigations tried to seek the primary origin of the thixotropic deformation behaviour from the corresponding microstructure through the semi-solid compression test. Shang et al.¹⁷ proposed that four deformation mechanisms coexisted during the thixotropic compression process of 6061 alloy, and the predominant mechanism depended on the compression temperature and strain rate. The deformation mechanisms included liquid flow, flow of liquid incorporating solid particles, sliding among solid particles and plastic deformation of solid particles. Hassas-Irani et al.¹⁸ explored the influence of initial microstructures on the thixotropic behaviour of A356 alloy during semi-solid compression and pointed out that due to the flow of liquid incorporating solid particle mechanism, globular microstructure exhibits the lowest flow resistance, followed by coarsened microstructure, rosette-like microstructure and cast microstructure. Some studies just concentrated on the local microstructure such as the centre microstructures and the edge microstructures during the compression test. However, the microstructures are greatly affected by the deformation zones it located. The detailed partition of the deformation zone is also less presented in the literature.

In this work, radial forging (RF) is applied to the strain-induced step in the SIMA process to obtain the semi-solid microstructure of A356.2 alloy, which can produce relatively large billet with large strain and good deformation uniformity. Then, high-solid fraction semi-solid compression tests at different compression temperatures and velocities are conducted to investigate the deformation behaviour of A356.2 alloy. Finally, four deformation zones are divided, and the influences of compression parameters on the microstructures at different deformation zones are investigated. All these works may lay an experimental foundation for the thixoforming of the semi-solid A356.2 alloy.

Experiment

Experimental material

Commercial as-cast A356.2 aluminium alloy is used as experimental material; its chemical composition and thermal characteristics are shown in Table 1. The semi-solid temperature range of A356.2 aluminium alloy is obtained by differential scanning calorimetry (DSC) curve at 10 K/min as shown in Figure 1(a). Figure 1(b) shows the liquid fraction versus temperature, which is obtained by integrating under the DSC curve. It can be seen that for temperatures of 833, 838 and 843 K, the corresponding liquid fractions were 4.9%, 7.8% and 12%, respectively. These belong to high solid fraction and are suitable for thixoforming.

Table 1.

Chemical composition (wt.%) and thermal characteristics of A356.2 alloy.

Si	Fe	Cu	Mg	Mn	Zn	Ti	Al	T_S (K)	T_L (K)
7.1	0.11	0.007	0.35	0.011	0.012	0.11	Balance	823	888

Figure 1.

(a) DSC curve (10 K/min) of the A356.2 aluminium alloy and (b) liquid fraction versus temperature derived from the DSC curve.

Semi-solid billet preparation and compression test

RF is introduced into the SIMA process for the first time, and its working principle is shown in Figure 2. First, the as-cast A356.2 alloy bar (Φ120 mm) was held at 573 K for 90 min. Then, the preheated aluminium bar was deformed via repeating RF process until the reduction in area reached 60% (Φ76 mm) and 75% (Φ60 mm), respectively. Finally, the forged bar was quenched subsequently. During the partial re-melting process, the forged bar was isothermally held at 853 K for 15 min and water quenched immediately to retain the semi-solid microstructure.

Figure 2.

Schematic diagram for the radial forging of the as-cast A356.2 alloy bar.

Moreover, compression specimens with dimensions Φ8 × 10 mm² were fabricated along the forging direction at the Φ60 mm forged bar. Then, the semi-solid compression tests at different isothermal holding temperatures (T_h) and compression velocities (V_c) were carried out by the Instron universal testing machine with a resistance furnace as shown in Figure 3. Three thermocouples were used to detect the temperature at the upper, middle and lower positions near the specimen before compression to retain the temperature in the furnace as uniform as possible. After compression, all the specimens were water quenched to retain the deformed microstructure. The detailed experimental matrix is given in Table 2.

Figure 3.

Schematic diagram for the compression test of the semi-solid A356.2 alloy.

Table 2.

Experimental parameters of the semi-solid compression for A356.2 alloy.

Parameters	Value
Holding temperature (T_h, K)	833, 838, 843
Holding time (t_h, min)	5, 10, 15, 20
Compression velocity (V_c, mm/min)	6, 30, 60
Compression ratio	55%

All samples were ground, polished and etched with an aqueous solution of 0.5% HF for microstructural observation. Microstructural observation was conducted on an OLYMPUS GX5 optical microscope. The average particle size (APS) and shape factor (SF) were calculated by equations (1) and (2), respectively. In these equations, A_N, N and P_N are area, the number and perimeter of solid grains, respectively. SF is a value that varies from 0 (for acicular particles) to 1 (for perfectly round sphere)

APS = \frac{\sum_{N = 1}^{N} \sqrt{4 A_{N} / π}}{N}

(1)

SF = \frac{\sum_{N = 1}^{N} 4 π A_{N} / P_{N}^{2}}{N}

(2)

Results and discussion

Microstructures of the semi-solid A356.2 alloy prepared by SIMA

As shown in Figure 4, the representative microstructure of as-cast A356.2 alloy consisted of primary α-Al dendrites and some inter-dendritic eutectic structures among them. Owing to the severe plastic deformation by RF, the dendrites turned to textured structures along the forging direction as shown in Figure 5. It can be seen from Figure 5(a) that most large α-Al dendrites were crashed and elongated obviously along the forging direction and mixed with the eutectic structures, forming the laminar structures as the reduction in area reached 60%. When the reduction in area further increased to 75%, the dendrites were further broken up and the laminar structures became so compact that some further elongated grains were superimposed with each other and formed some aggregations as shown in Figure 5(b). Figure 6 shows the microstructures of A356.2 alloy with different reductions of area after isothermal holding at 853 K for 15 min. As shown in Figure 6(a), when the reduction in area was 60%, there existed many solid particles with larger size, although the particles have obviously undergone spheroidization. With further increasing reduction in area, the solid particles became finer and more spheroidal as shown in Figure 6(b). Liquid pockets were trapped in some large spheroidal grains in Figure 6(a) and (b) because the adjacent elongated particles that have perfect matching crystallographic orientation coalesced during the partial re-melting process, forming the new solid particles containing the eutectic in them. The APS and SF of the solid particles in 60% and 75% reduction in area were about 87 and 83 µm and 0.72 and 0.75, respectively. It can be seen that increasing the reduction in area refined the particle size and promoted the degree of spheroidization, but its variation was not obvious. This is because the induced strain introduced by RF may be high enough to produce good enough semi-solid microstructure as the reduction in area was 60%. Higher reduction in area can provide more distortion energy that stored in the vacancies, lattice defect and dislocation multiplication. During reheating process, vacancies will combine and dislocations will climb and cross slip to decrease the free energy, which results in the occurrence of recovery and recrystallization.¹⁹ Thus, more distortion energy would provide larger driving force to recrystallize finer grains during reheating. In addition, more grain boundary and sub-grain boundary areas were generated with increasing reduction in area. This led to greater potentiality for the development of recrystallization nuclei.²⁰ Thus, the increased nucleation rate can also give finer recrystallized grains. Finally, high-angle boundaries of more recrystallized grains were penetrated by liquid, leading to the entire fragmentation of large particles to smaller and more spherical ones. However, it can be seen from Figure 6 that the distortion energy may have reached a peak value when the reduction in area was 60% since further increase in the reduction in area to 75% cannot decrease the APS obviously. This is possibly due to that the density of vacancies, lattice defect and dislocation may be saturated as the reduction in area reached 60%, and some of them would be neutralized with further increased reduction in area to 75%. Such two dislocations with opposite Burgers vectors will neutralize each other.²¹ Therefore, the total amount of distortion energy did not increase obviously. Furthermore, higher reduction in area induced that the forged alloy was more thermodynamically unstable and owned stronger atomic diffusion capacity during reheating process. This will further improve the degree of spheroidization at higher reduction in area because spheroidization is a diffusion-controlled process.

Figure 4.

Microstructures of as-cast A356.2 aluminium alloy.

Figure 5.

Microstructures of radial forged A356.2 aluminium alloy with different reductions in area: (a) 60% and (b) 75%.

Figure 6.

Microstructures of radial forged A356.2 alloy with different reductions in area after isothermal holding at 853 K for 10 min: (a) 60% and (b) 75%.

Characteristics of stress–strain curves and microstructures at different deformation zones

Figure 7 shows typical stress–true strain curves at various isothermal holding times (t_h) when T_h was 833 K and V_c was 6 mm/min. It can be seen that the whole stress–strain curve gradually decreased with the increasing t_h, but the descend range was limited. This may attribute to the slight improvement of the liquid film distribution around the solid particles to reduce the friction with longer t_h. However, there was no obvious difference on the microstructures. Considering the above-mentioned reasons, we finally chose 10 min as the isothermal holding time in the following experiments. Furthermore, the variation trend of these curves can be divided into three basic stages: strain hardening, thixotropic strain softening and steady deformation stage.²²

Figure 7.

Stress–strain curves of the semi-solid A356.2 alloy under different isothermal holding times (T_h = 833 K and V_c = 6 mm/min).

The semi-solid A356.2 alloy can be viewed as a solid skeleton that is filled with liquid which can possess certain static pressure. As the deformation was applied initially, the flow stress increased rapidly and reached a peak value, which was designated as strain hardening process. With the increase in local deformation, the solid skeleton began to bend and fell apart. When the thixotropic collapse occurred, the flow stress decreased. This was called thixotropic strain softening. Because both T_h and V_c were relatively low, the surface of the specimen was not ruptured obviously, and little liquid was extruded outside the surface; thus, the thixotropic strain softening process maintained a very short period.²³ At steady deformation stage, the stress–strain curve did not maintain a certain value just as in the compression tests with high strain rate in high liquid fraction sample²⁴ but presented a steady and slow increase. Although the liquid film reduced the friction among the solid particles, there still existed many solid particles that were contacted with each other and compressed seriously to adapt the deformation in the centre of the specimen due to the lower T_h. Even worse, the densification of particles, in turn, further increases the friction and prohibits the sliding and rotation of the particles. Therefore, the flow stress was increased gradually with the increased strain.

Figure 8 shows a quarter of the semi-solid compression sample and the corresponding microstructures at different deformation zones. Based on the different deformation characteristics of the microstructures, the compression sample can be divided into four deformation zones as shown in Figure 8: zone 1: hard deformation zone, zone 2: transition deformation zone, zone 3: severe deformation zone and zone 4: free deformation zone. In the following texts or figures, zone 1, zone 2, zone 3 and zone 4 represented the microstructures in different deformation zones, respectively, just as defined above. In zone 1, all particles still maintained the solid skeleton, but the particles at the bottom were bigger than those on the top because they were slightly stretched along the compression direction. In zone 2, the liquid incorporating solid particles flowed to the edge obviously, and the solid particles slid or rotated to adapt the liquid flow. Some large clusters were formed by several elongated grains due to the shearing force generated by the slip between the solid particles. In zone 3, the solid skeleton was completely destroyed, and the orientation of all grains turned to be vertical to the compression direction. Most elongated grains were further compressed into slender grains and formed some large aggregations with fuzzy grain boundaries. Some residual liquid was trapped among them. In zone 4, most grains had no obvious deformation and distributed more uniformly. Although the grains slightly flow to the edge with the liquid, more liquid in zone 4 than other zones ensured the free sliding and rotation of the solid grains. Therefore, the grains almost retained their morphology. From the above, it can be seen that the unbroken solid skeleton in zone 1 and the squashed grains in zone 3 may be attributed to the increase in flow stress, while the mixed flow of the liquid and solid particles in zone (2) and free sliding and rotation of grains in zone (4) may help in the thixotropic decrease in the flow stress. Therefore, we can conclude that liquid flow, mixed flow of the liquid and solid particles, sliding, rotation and plastic deformation of solid particles coexisted in the specimen during the compression process. However, the predominant deformation mechanism in different zones may be quite different. These differences not only affected the morphology of the grains in different deformation zones but also codetermined the magnitude and trend of the flow stress.

Figure 8.

Microstructures of the semi-solid A356.2 at different deformation zones (T_h = 560 °C, t_h = 10 min and V_c = 6 mm/min).

Effect of V_c and T_h on the stress–strain curves

Figure 9 shows a series of stress–strain curves under different V_c and T_h. It could be seen that the developing trends of the curves resembled basically. The flow stresses were all obviously increased with increasing V_c but inconspicuously decreased with increasing T_h. This phenomenon proved that in the low-liquid fraction compression tests, the flow stress was very sensitive to the V_c but insensitive to the T_h. In order to further investigate the influence of average strain rate ( $\bar{\overset{\cdot}{ε}}$ ) on the peak stress (σ_p), the power law $σ_{p} = K_{c} {\bar{\overset{\cdot}{ε}}}^{m}$ was adopted, where $σ_{p}$ is the peak stress before thixotropic collapse, $\bar{\overset{\cdot}{ε}}$ is the average strain rate that was calculated from the V_c, $K_{c}$ is the constant reflecting the effect of material structure and temperature and m is the strain rate sensitivity, which is the slope of the curve $d (\ln σ_{p}) / d (\ln \bar{\overset{\cdot}{ε}})$ . The double logarithmic relationship between the σ_p and $\bar{\overset{\cdot}{ε}}$ at each temperature level was plotted in Figure 10(a). It can be seen from Figure 10(a) that within the investigated temperature range, the relationships between the $σ_{p}$ and $\bar{\overset{\cdot}{ε}}$ in the double logarithm coordinates were fairly linear. The value of m and the correlation coefficient (R²) at all temperature levels is presented in Table 3. This indicated that in the semi-solid temperature range, the rate-dependent behaviour of the A356.2 alloy obeyed the power law. Moreover, it is well known that the quantitative relationship between the σ_p and the T_h at the same strain rate can be expressed by mQ/R = d(lnσ_p)/d(1/T), where m is the strain rate sensitivity, Q is the apparent activation energy, R is the gas constant, σ_p is the peak flow stress and T is the absolute temperature. Figure 10(b) illustrated the relationship between the logarithm of σ_p and the reciprocal temperature (1/T_h) at different $\bar{\overset{\cdot}{ε}}$ . The linear correlation coefficients (R²) of lnσ_p and (1/T) are shown in Table 4. Furthermore, applying the $\bar{m}$ at different temperature ranges as given in Table 3, the Q at different $\bar{\overset{\cdot}{ε}}$ was calculated and also presented in Table 4. The correlation coefficients (R²) at each strain rate in Table 4 also revealed a well linear behaviour of the experimental data. However the d(lnσ_p)/d(1/T) at 0.1 s⁻¹ was a little abnormal, which might be considered as the consequence of the inherent mutability of compression in the semi-solid state.²⁵

Figure 9.

Stress–strain curves of the semi-solid A356.2 alloy under different V_c: (a) T_h = 833 K, (b) T_h = 838 K and (c) T_h = 843 K.

Figure 10.

Relationship between the lnσ_p and (a) the $\ln \bar{\overset{\cdot}{ε}}$ at different T_h and (b) the T⁻¹ at different $\bar{\overset{\cdot}{ε}}$ .

Table 3.

Values of strain rate sensitivity in the semi-solid state.

Compressive state	Semi-solid state
Temperature (K)	833	838	843
m	0.2719	0.2810	0.2742
R ²	0.9988	0.999	0.9975
$\bar{m}$	0.2757

Table 4.

Values of the apparent activation energy in the semi-solid state.

Compressive state	Semi-solid state
$\bar{\overset{\cdot}{ε}}$ (s⁻¹)	0.01	0.05	0.1
d(lnσ_p)/d(1/T)	8512	8974	7918
R ²	0.9942	0.999	0.9988
Q	255.4	269.2	238.8
$\bar{Q}$ (kJ/mol)	254.5

Effect of V_c and T_h on the microstructures at different zones

Figure 11 presents the microstructures at different zones for different V_c when T_h is 838 K. It can be seen that at zone 1, as V_c increased, the globular grains on the top of the metallography still retained their morphology, but more and more grains at the bottom were elongated and coalesced by the increased shearing stress generated by the increasing V_c. At zone 2, the number and size of the elongated grains that flowed along the liquid flow direction increased sharply with the increasing V_c. Especially, when V_c = 60 mm/min, more grains were contacted with each other and merged into many large elongated grains along the liquid flow direction. At zone 3, higher V_c induced larger and more distorted grains that were vertical to the compression direction. Most of the liquid in zone 3 flowed away. The remaining liquid had no enough time to coordinate the sliding and rotation of the grains at higher compression velocity. Therefore, more elongated grains were compressed and merged with each other, forming the larger grains with some trapped liquid as V_c increased. However, the grains in zone 4 had no severe plastic deformation but grew up gradually with the increasing V_c. High V_c may enhance the liquid segregation, so more liquid existed at the solid–liquid interfaces of the grains at zone 4, which contributed to the Ostwald ripening of the solid particles. This phenomenon, on one hand, benefited sliding and rotation of the particles to adapt the macroscopical deformation; on the other hand, it may promote the coarsening of some coalesced grains, forming these large grains at zone 4 when V_c = 60 mm/min. Figure 12 shows the corresponding variations of APS and SF, respectively. The APS increased from zone 1 and reached the maximum at zone 3 and then decreased at zone 4 for different V_c, while the opposite was the case for SF as shown in Figure 12(b). Furthermore, with the increasing V_c, the APS was increased, while the SF was decreased, respectively, at each zone. This may be due to higher deformation degree and compression velocity that could lead to more grains contacted with each other and merged into larger and more irregular grains as shown in Figure 11. Furthermore, the abnormal increase in SF at zone 4 when V_c = 60 mm/min may be mainly attributed to the coalescence and Ostwald ripening of the solid particles as mentioned above. From the above, it can be concluded that the compression velocity greatly influenced the size and morphology of particles at different zones.

Figure 11.

Microstructures under different V_c at different deformation zones (T_h = 838 K).

Figure 12.

Variations of (a) APS and (b) SF of microstructures under different V_c at different deformation zones (T_h = 838 K).

Figure 13 shows the microstructures of the semi-solid A356.2 alloy under different T_h at different deformation zones, when V_c is 30 mm/min. As shown in Figure 13, when T_h was 833 K, most globular particles stick to each other and formed irregular grains at zone 1 because there was no enough liquid to penetrate the grain boundaries. As T_h further increased, the liquid film was thicker, and the particles were more global and distributed more homogeneous at zone 1. Meanwhile, the number of elongated grains decreased. At zone 2, most grains were bended along the liquid flow direction. Because of lack of fluid lubrication, the bended grains contacted with each other and formed some large aggregations at 833 K. However, the number of the aggregations was gradually decreased due to the thick liquid film as T_h increased. At zone 3, the grain boundary became unapparent, and most elongated particles were transformed into large and spindly grains. But as T_h increased, more liquid contributed to separate the aggregations and improve the morphology of the particles. Meanwhile, some liquid was trapped between the incorporative squashed grains, forming some liquid pockets at higher T_h. Zone 4comprises many fine and spherical particles on account of more liquid than other zones. The particles became more isolated with increasing T_h. However, the liquid flow to the edge resulted in many elongated particles along the liquid flow direction at 843 K, which deteriorated the morphology of the particles. Figure 14 shows the variations of APS and SF of the semi-solid A356.2 alloy under different T_h at different deformation zones. As shown in Figure 14(a), the APS was increased from zone 1 to zone 3 and then decreased at zone 4 at different T_h. Furthermore, the APS was generally decreased with the increasing T_h at different zones because more liquid helped to moderate the extent of aggregations. On the contrary, the SF was decreased from zone 1 to zone 3 and then increased at zone 4 at different T_h. Meanwhile, the SF was generally increased with T_h at different zones because more liquid can penetrate the boundary of the contacted particles and contributed to the spheroidization of the independent particles. But the APS at zone 3 and SF at zone 4 at 843 K were a little abnormal. This phenomenon may be due to the drastic liquid flow from zone 3 to zone 4 caused by the instability of the semi-solid billet at higher temperature. On one hand, this drastic liquid flow to the edge resulted in further connection of the compressed particles at zone 3 seriously, forming some large and twisty grains, which led to the abnormal increase in APS at zone 3; on the other hand, this drastic liquid flow incorporating solid particles into zone 4 induced the forced motion and elongation of the particles in zone 4, which deteriorated the spheroidization degree of the particles in zone 4. It can be seen from above that the compression temperature mainly affected the fraction and distribution of the liquid in different deformation zones and then affected the size and morphology of the grains.

Figure 13.

Microstructures under different T_h at different deformation zones (V_c = 30 mm/min).

Figure 14.

Variations of (a) APS and (b) SF of microstructures under different T_h at different deformation zones (V_c = 30 mm/min).

Conclusion

RF can induce large strain to turn α-Al dendrites into compact textured structures at 60%–75% reduction of area, and the microstructure can be transformed into fine and spherical particles during isothermal holding at 853 K for 15 min. However, the APS and SF only improved slightly at 75% reduction in area compared with that at 60% because the distortion energy tended to be saturated at 60%.

During the thixotropic compression process, the compression sample can be divided into hard deformation zone, transition deformation zone, severe deformation zone and free deformation zone based on the deformation characteristic of the microstructures. Although liquid flow, mixed flow of the liquid and solid particles, sliding, rotation and plastic deformation of solid particles coexisted in the specimen, different zones have different predominant deformation mechanisms which affected both the morphology of the grains and the variation trend of the flow stress.

The flow stress were obviously increased with increasing V_c but inconspicuously decreased with increasing T_h, proving that the flow stress was sensitive to the V_c but insensitive to the T_h. The rate-dependent deformation behaviour of the A356.2 alloy obeyed the power law $σ_{p} = K_{c} {\bar{\overset{\cdot}{ε}}}^{m}$ . The average strain rate sensitivity was 0.2757, and the average apparent activation energy was 254.5 kJ/mol.

With increasing V_c and T_h, the APS was increased from hard deformation zone and reached the maximum at severe deformation zone and then decreased at free deformation zone, while the opposite was the case for SF. At each zone, the APS was increased while the SF was decreased, respectively, with the increasing V_c because higher V_c caused more grains to merge into larger and more irregular grains, while the opposite was the case for the increasing T_h because higher T_h generated more liquid which helped to isolate the contacted particles and improve their spheroidization degree.

Footnotes

Acknowledgements

We thank Mr Guanhai Yan for providing relevant suggestions and performing language corrections.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Science Foundation of China (grant no. 51335009) and Major National Science and Technology Project of China (grant no. 2012ZX04004021).

References

Yekta

Vanini

SAS

Pouyafar

. Feasibility of partial remelting process in globular microstructure preparation for semi-solid processing of M2 high-speed steel. Proc IMechE, Part B: J Engineering Manufacture 2015; 229: 1002–1009.

Yekta

Vanini

SAS

. Flow behaviour and rheological properties of M2 high-speed steel in semi-solid state. Proc IMechE, Part B: J Engineering Manufacture 2014; 228: 1611–1620.

Lee

Kang

. Effect of solid fraction on formability and mechanical properties in a vertical-type rheo squeeze-casting process. Proc IMechE, Part B: J Engineering Manufacture 2011; 225: 184–196.

Jang

Jin

Bolouri

et al . Effect of forming conditions on mechanical properties of rheoformed thin plates with microchannels using electromagnetic stirring. Proc IMechE, Part B: J Engineering Manufacture 2014; 228: 399–412.

Wen

Wang

Yang

et al . On a reshaping method of clinched joints to reduce the protrusion height. Int J Adv Manuf Tech 2014; 71: 1709–1715.

Mirjavadi

Alipour

Hamouda

AMS

et al . Investigation of the effect of Al-8B master alloy and strain-induced melt activation process on dry sliding wear behavior of an Al–Zn–Mg–Cu alloy. Mater Design 2014; 53: 308–316.

Bolouri

Shahmiri

Kang

. Study on the effects of the compression ratio and mushy zone heating on the thixotropic microstructure of AA 7075 aluminum alloy via SIMA process. J Alloy Compd 2011; 509: 402–408.

Alipour

Emamy

Ebrahimi

SHS

et al . Effects of pre-deformation and heat treatment conditions in the SIMA process on properties of an Al–Zn–Mg–Cu alloy modified by Al–8B grain refiner. Mat Sci Eng A: Struct 2011; 528: 4482–4490.

Moradi

Nili-Ahmadabadi

Poorganji

et al . Recrystallization behavior of ECAPed A356 alloy at semi-solid reheating temperature. Mat Sci Eng A: Struct 2010; 527: 4113–4121.

10.

Tao

Jiang

Chen

et al . Application of cyclic upsetting-extrusion to semi-solid processing of AZ91D magnesium alloy. T Nonferr Metal Soc 2013; 23: 909–915.

11.

Parshizfard

Shabestari

. An investigation on the microstructural evolution and mechanical properties of A380 aluminum alloy during SIMA process. J Alloy Compd 2011; 509: 9654–9658.

12.

Yan

Zhao

et al . Microstructural evolution of A356.2 alloy prepared by the SIMA process. Mater Charact 2012; 69: 45–51.

13.

Jiang

Chen

Wang

et al . Microstructure in the semi-solid state and mechanical properties of AZ80 magnesium alloy reheated from the as-cast and extruded states. Acta Metall Sin 2013; 26: 473–482.

14.

Haghdadi

Zarei-Hanzaki

Heshmati-Manesh

et al . The semisolid microstructural evolution of a severely deformed A356 aluminum alloy. Mater Design 2013; 49: 878–887.

15.

Zhao

Chen

Tang

et al . Microstructural evolution and tensile mechanical properties of AM60B magnesium alloy prepared by the SIMA route. J Alloy Compd 2010; 497: 402–411.

16.

Wang

Mei

et al . Microstructure evolution and grain coarsening behaviour during partial remelting of cyclic extrusion compression formed AZ61 magnesium alloy. Acta Metall Sin 2013; 26: 149–156.

17.

Shang

Tang

et al . Deformation mechanism and forming properties of 6061Al alloys during compression in semi-solid state. T Nonferr Metal Soc 2010; 20: 1725–1730.

18.

Hassas-Irani

Zarei-Hanzaki

Bazaz

et al . Microstructure evolution and semi-solid deformation behavior of an A356 aluminum alloy processed by strain induced melt activated method. Mater Design 2013; 46: 579–587.

19.

Wang

et al . Semisolid microstructure evolution of the predeformed AZ91D alloy during heat treatment. J Alloy Compd 2005; 395: 108–112.

20.

Atkinson

Burke

Vaneetveld

. Recrystallisation in the semi-solid state in 7075 aluminium alloy. Mat Sci Eng A: Struct 2008; 490: 266–276.

21.

Bolouri

Shahmiri

Cheshmeh

. Microstructural evolution during semisolid state strain induced melt activation process of aluminum 7075 alloy. T Nonferr Metal Soc 2010; 20: 1663–1671.

22.

Wang

Phillion

. Development of a visco-plastic constitutive modeling for thixoforming of AA6061 in semi-solid state. J Alloy Compd 2014; 609: 290–295.

23.

Hong

Wang

J-J

. Thixotropic compression deformation behavior of SiC_p/AZ61 magnesium matrix composites. T Nonferr Metal Soc 2010; 20: s811–s814.

24.

Favier

Atkinson

. Micromechanical modelling of the elastic–viscoplastic response of metallic alloys under rapid compression in the semi-solid state. Acta Mater 2011; 59: 1271–1280.

25.

Wang

Zhou

et al . Constitutive behavior of C_sf/AZ91D composites compressed at elevated temperature and containing a small fraction of liquid. Compos Sci Technol 2011; 71: 955–961.

Deformation behaviour and microstructures of semi-solid A356.2 alloy prepared by radial forging process during high solid fraction compression

Abstract

Keywords

Introduction

Experiment

Experimental material

Semi-solid billet preparation and compression test

Results and discussion

Microstructures of the semi-solid A356.2 alloy prepared by SIMA

Characteristics of stress–strain curves and microstructures at different deformation zones

Effect of Vc and Th on the stress–strain curves

Effect of Vc and Th on the microstructures at different zones

Conclusion

Footnotes

Acknowledgements

Declaration of conflicting interests

Funding

References

Effect of V_c and T_h on the stress–strain curves

Effect of V_c and T_h on the microstructures at different zones